KR950011198B1 - Television communicaiton method and apparatus - Google Patents

Abstract

No content.

Description

TV information communication method and apparatus

1 is a block diagram of a transmitter implementing the principles of the present invention in a four-dimensional channel mapping scheme for HDTV.

2 is a block diagram of another transmitter implementing the principles of the present invention in a two-dimensional channel mapping scheme for HDTV (the scheme includes lattice coding).

3 is a block diagram of a receiver receiving a transmission signal transmitted to the transmitter of FIG.

4-11 are signal coordinate diagrams useful for illustrating the principles of the present invention.

* Explanation of symbols for main parts of the drawings

101, 201: TV signal source 104,204: signal source coder

121: Four-dimensional mapper 215: Quadrature phase grating coder

216: in-phase lattice coder 221: quadrature phase one-dimensional mapper

222: In-Phase One-Dimensional Mapper

[Background of invention]

The present invention relates to, but is not limited to, transmission of digital data, and includes transmission of digital data representing a television (TV) signal.

It is generally agreed that some form of digital transmission will be required for next generation television (TV) technology, commonly referred to as high definition television or HDTV. This condition is mainly due to the fact that much more powerful video compression schemes can be performed with digital signal processing than analog signal processing. However, since the positional sensitivity of digital transmission is sensitive to small changes in signal-to-noise ratio or SNR at various reception sites, there has been a concern about the transition to a fully digital transmission system.

This phenomenon, sometimes referred to as the "threshold effect", can be accounted for in the case of two TV receivers located 50 and 63 miles, respectively, from a TV station. Since the strength of the broadcast signal is approximately inversely proportional to the square of the distance, it is easily demonstrated that the difference in signal strength received by the TV receiver is about 2 dB. Now assume that a digital transmission scheme is used and transmission to a receiver 50 miles away represents a bit error rate of 10 ⁻⁶ . If the additional signal loss of 2 dB for another TV receiver translates into a 2 dB reduction in SNR at the input of the receiver, the receiver will operate at a bit error rate of about 10 ⁻⁴ . This kind of bit error rate results in very good reception for TV receivers 50 miles away while the other TV receiver has very poor reception. This rapid deterioration (catastrophic) over such short distances is generally considered unacceptable in the broadcast industry (in comparison, the degradation in current analog TV transmission schemes is much slower).

Therefore, there is a need for a digital transmission scheme that can overcome these problems and be used in the TV field. Solutions used in other digital transmission environments include the use of a) regenerative repeaters in cabled transmission systems or a) the use of fall-back data rates or improved telephone lines in the field of voiceband data. It is not applicable at all to the free space broadcasting environment.

[Overview of invention]

The core recognition of the present invention is that certain features of the prior art digital transmission systems are inadequate when used in a TV transmission environment, for example, and that is a difficult problem. In particular, digital transmission systems have been traditionally designed to protect approximately equal amounts of damage from the generally bit-mode data elements transmitted on a communication channel. This approach is desirable when the digital transmission mechanism is transparent to the user's data and does not know the content of the data, for example in the voice band data or digital radio field. However, if all the bits are treated the same, then by changing the channel conditions, everyone will be affected as well, thus causing catastrophic as described in the above example.

According to the present invention, the disadvantages of standard digital transmission for over-the-air broadcasting of digital TV signals include some form of source coding followed by some form of channel mapping. The channel mapping is referred to herein as Catastrophe-Resistant (CR) mapping.

In particular, the signal source coding step causes the TV signal to be represented by two or more data flows, but in the channel mapping step, the mapping is such that data elements of multiple data flows are unlikely to be error detected at the receiver. In a preferred embodiment, the first data flow of the above-described data flow carries a component which is considered to be the most important of the total TV signals, as will be discussed in detail herein below, which data flow has the lowest error in the data element. Mapped to have the possibility. The second data flow in the data flow carries a component of the total TV signal that is considered less important than the component of the first data flow, where the data element of the data flow is not as low as the error probability of the first data flow. Mapped to have the possibility. In general, the entire TV signal can be represented by any number of data flows, each carrying a component of differing importance and each having a respective probability of error. This method increases bit error rate at the receiver as the distance from the broadcast transmitter increases, so that the first to be affected is the bits representing the less important TV signal information, so that the degradation of the reception picture quality at the TV receiver location is good. do.

The present invention is not only limited to TV signals, but may actually be used in any environment where it is desirable to provide different levels of error protection for different components of the information communicated.

[details]

Before giving a detailed description of the transmitters of FIGS. 1 and 2 and the receivers of FIG. 3, it would be helpful to first consider the theoretical background of the present invention.

The various digital signal processing concepts described herein (except, of course, the inventive concept itself) are well known in the digital wireless and voiceband data transmission (modem) technology and need not be described in detail herein. They are multidimensional signal processing using N-dimensional signal constellation (where N is an integer), trellis coding, scrambling, passband shaping, equalization, Viterbi (Viterbi) or maximum extreme decoding and the like. This concept is described in US Patent No. 3,810,021, issued May 7, 1974 to I. Kalet et al .; US Patent No. 4,015,222, issued March 29, 1977 to J. Werner; US Patent No. 4,170,764, issued to J. Salz et al. On 1979.10.9; US Patent No. 4,247,940, issued January 27, 1981 to K. H. Mueller et al .; US Patent No. 4,304,962, issued December 8, 1981 to R. D. Fracassi et al .; To Mr. A. Gersho et al .; 1984. 6. US Pat. No. 4,457,004, issued 26 characters; US Patent No. 4,489,418, issued December 18, 1984 to J. E. Mazo; US Patent No. 4,520,490, issued 28 May 1985 to L. Wei; G. D. Forney, Jr. Was described in US patents, such as US Pat. No. 4,597,090, issued June 24, 1986, all of which are incorporated herein by reference.

Referring to the drawings, FIG. 4 shows standard two-dimensional data transmission coordinates of the type commonly used in digital wireless and voiceband data transmission systems. In this standard scheme, commonly referred to as quadrature amplitude modulation (QAM), a data word each containing four bits is mapped to one of sixteen possible two-dimensional signal points. The coordinates are therefore called "standard 16QAM". Each signal point has an in-phase or I coordinate on the abscissa and a quadrature phase or Q coordinate on the ordinate. Note that the coordinates of the signal points on each axis are ± 1 or ± 3, so the distance between each point and each point vertically or horizontally adjacent to the point is "2", so that all points are the same (dataword to a specific signal point). The process of mapping is referred to herein as "channel mapping" and signal points are sometimes referred to as "channel symbols").

Now look at the 16 point coordinates of Figure 5 embodying the principles of the present invention. The difference between the coordinates and the coordinates of FIG. 4 is the relative distance between different signal points.

In particular, since the distances between all adjacent points in FIG. 4 are the same, the same probability of error is necessarily provided for every bit represented by the signal point. (Transmission errors are caused by noise, phase jitter, and various other channel phenomenon damages. As a result, it occurs when the transmitted signal point is moved to some extent from the original position of the coordinates so that the transmitted signal point appears at the receiver to which the different signal point is transmitted). On the other hand, the distance between adjacent points in FIG. 5 is not the same at all points. In particular, the minimum distance between points in one particular quadrant of FIG. And the minimum distance between points in adjacent quadrants is twice that to be. Thus, the likelihood of error in identifying in which quadrant the transmission point was located at the receiver is less than the likelihood of error in identifying which point in that quadrant was the actual point. The first flow pair from a point in the second quadrant that represents the minimum distance between the signal points representing different values of the data elements of the first data flow (e.g., first-stream dibit 01). The minimum distance between points in the first quadrant representing bit 01 ( ) Is the minimum distance between signal points representing different values of the data elements of the second data flow (e.g., the point in the first quadrant representing the second flow pair bit (00) and the second flow pair bit (01). The minimum distance between points in the same quadrant Due to greater than

Since two bits out of four bits of each transmitted data word are more important than the other two bits, it is assumed that more error protection is required than the other two bits. This uses two more significant bits to select one of the four quadrants, as indicated by the double-bits encircled in FIG. 5, according to the present invention, and as indicated by the double-bits adjacent to each point. This is done by using two different bits to select one of the four points in each quadrant. Since the likelihood of misidentifying the quadrant of the transmitted signal point is lower than that of misidentifying the signal point itself, the desired protection is achieved.

More generally speaking, the coordinates are divided into groups of signal points, each group being divided into subgroups each consisting of one or more signal points. At least one data element, eg a bit, from each data word to be mapped identifies a group from which a signal point representing the data word is to be derived, and at least one other data element identifies a subgroup within that group. If the subgroup contains one or more signal points, another data element is used to finally identify a particular one of these signal points (for this purpose, the subgroup may be further divided into sub-subgroups).

According to the invention, a) the groups and subgroups are arranged such that the receiver is less likely to misjudge the group from which the transmitted signal point originated, less than the probability of misrecognizing the subgroup from which the signal point originated, and b) the data identifying the group. The elements are arranged to display more important information than the information represented by the data elements identifying the subgroups.

A general variation of the coordinates of FIG. 5 is shown in FIG. 6, where the coordinate values are Instead ± α and ± β. The coordinates are not limited to any particular size, ie the number of signal points. For example, the upper right quadrant of the standard 64QAM coordinates is shown in FIG. 7, and a general 64 point coordinate is shown in FIG. 8 that embodies the principles of the present invention and provides three different levels of protection.

Before continuing, it is useful to define some expressions. As mentioned above, channel mapping according to the present invention is referred to herein as catastrophic resistance (CR) mapping. In general, (n _1, n _2, ... n _k; m) CR mapping is to provide a level of the first (top) protecting the n ₁ bits, service level, etc. of the second protected sequentially to n _2-bit It will be a mapping. The final entry of the mapping identification is the remainder of the total number of information bits transmitted, i.e. m = n ₁ + n ₂ +... + n _k . By this definition, each CR mapping shown in FIGS. 5 and 6 is a (2, 2; 4) mapping. FIG. 8 shows only the upper right quadrant as an example of the 64 point (2, 2, 2; 6) mapping, and an example of the 16 point (1, 2, 1; 4) mapping is shown in FIG. Finally, note that the standard QAM mapping (m; m) of the form shown in FIGS. 4 and 7 can be considered a CR mapping.

The types of trade-offs possible in the design of C-R mapping are briefly described. First, it is assumed that the strength of the transmitted signal is subject to the average strength limitation. Suppose ai and bi represent I and Q discrete signal point levels, respectively, and these signal points are independent of each other. The average intensity constraint then requires the following equation for all signal point level schemes under consideration.

Now assume that SNR _ni represents the amount of SNR needed to achieve a particular performance for a bit with a first level of protection. The variation in the amount of SNR these bits require to achieve this level of performance compared to standard (m; m) mapping is then defined as follows.

ΔSNR _ni ≡SNR _ni -SNR _m (2)

Where SNR _m is the amount of SNR that the (m; m) mapping needs to achieve the same performance (this mapping is a mapping that provides the same amount of protection for all bits). Using equations (1) and (2), we can obtain the following relationship to the mapping (2,2; 4) shown in FIG. 6;

(3)

Here, the SNR increment is expressed in dB. Using α in FIG. 6 as a parameter of equation (3), it is possible to first determine the β value and then to determine the SNR increment. Some calculations are given in Table I.

TABLE 1

Correlation to (2,2; 4) Mapping

To give some meaning to the entries in Table I, As is the case, consider a specific example corresponding to the signal coordinates shown in FIG. The SNR increment for the best-protected 2 bits is given in the third column. For the case under consideration, the SNR increment for the bit is -3 dB. Thus, for a given probability of error, these bits can tolerate an SNR that is 3 dB less than the SNR required for a standard 16 point QAM system. On the other hand, as can be seen in the fourth column, the least protected bits will require an SNR of more than 3 dB to achieve the same performance as a standard QAM system.

The correlation achieved in the above example may seem very strict; On the other hand, the noise sensitivity for the first two bits is reduced by 3 dB; On the other hand, this sensitivity is increased by 3dB for the other two bits. This simple correlation rarely occurs, as can be seen in the other entries in Table I. For example, for α = 1.2, the best protected bit gains robustness to greater noise than is lost by the least protected bit. This should be pursued in the design of efficient C-R mapping.

The invention is not limited to two-dimensional coordinates, but in fact can be implemented with N-dimensional coordinates where N≥2. In fact, increasing the number of dimensions gives greater flexibility in designing an efficient mapping. One way to perform multidimensional C-R mapping with a QAM system is to use different two-dimensional C-R mappings at successive signal point intervals. By way of example, a four-dimensional coordinate diagram may be used to map all possible two-dimensional signal points from the (2,2; 4) mapping of FIG. 5 from the (1,2,1; 4) mapping of FIG. Can be made by concatenating (connecting) all possible two-dimensional signal points

This mapping process can facilitate providing (3,2,3; 8) four-dimensional CR mapping. In particular, the maximum spacing between the points of the coordinates of FIG. This is the minimum distance between a point in one quadrant and a point in the other quadrant. The same maximum interval divides the coordinate diagram of FIG. 9 into an upper half and a lower half. Thus, the level of maximum protection is represented by two bits selecting a quadrant from the coordinates of FIG. 5 as indicated by the double bits surrounded by circles in FIG. 5 and by the double bits surrounded by the circles in FIG. 3 bits, which is a third bit, which selects one of the upper half and the lower half of the coordinate of the figure. The second largest gap is the distance between the columns of coordinates of FIG. 9, with a minimum of two.

Thus, the second highest level of protection is achieved for two bits that select one of the four columns from the coordinates of FIG. 9, as indicated by the double bits surrounded by the rectangle in FIG.

Finally, the minimum distance is the minimum distance between the points in a quadrant and the points in the column, in coordinates of FIG. Thus, the lowest level of protection is indicated by the next two bits adjacent to each point in FIG. 5, as indicated by two bits selecting the point in the selected quadrant of the coordinates of FIG. 5 and the next single bit adjacent to each point in FIG. It can be done again for 3 bits, the third bit selecting one of the two halves of the selected half of the coordinates of FIG. 9 and the selected column.

Thus, for example, an 8-bit word (01110100) will result in the selection of a four-dimensional signal point made up of the connection of point A in FIG. 5 and point A 'in FIG. In particular, the first and second bits 01 select the upper left quadrant of FIG. 5; The third bit 1 selects the lower half of FIG. Fourth and fifth bits 10 select the second column on the right side of FIG. 9; The sixth and seventh bits 10 select a point A from the quadrant of FIG. 5 previously selected; Eighth bit (0) selects point A 'in the half and column of Figure 9 previously selected.

For this mapping, the SNR condition for two bits with the second level of protection is the same as the SNR condition for the standard QAM signal coordinates of FIG. The highest protected 3 bits and the lowest protected 3 bits and the lowest protected 3 bits have an SNR condition derived as described above for two-dimensional (2, 2; 4) mapping.

The transmitter of FIG. 1 will now be described. The TV signal source 101 generates an analog TV signal that is applied to a signal source coder 104. The signal source coder generates a digital signal in which at least one subset of data elements represents information that is more important than the information represented by the rest of the data elements. Two examples of how this signal is generated are described herein below.

The source coded signal is CR mapped according to the present invention using the four-dimensional mapping described above, wherein each four-dimensional signal point is a two-dimensional signal point from the coordinates of FIG. 5 associated with the two-dimensional signal point from the arrangement of FIG. It is composed. In accordance with an aspect of the present invention, despite any processing of bits that may be needed before being mapped to a four-dimensional signal point by the four-dimensional mapper 121, the mapping of the bits to be given a particular level of protection by the mapping. It has been recognized that it is desirable to protect the identity. If this identity is not maintained, of course, it would be impossible to assign different levels of protection to the various data flows representing the TV signal.

In particular, in this embodiment, it is desirable to scramble the bits constituting the digital signal to ensure a relatively uniform energy distribution over the frequency band the signal takes. Thus, these bits are scrambled into three separate groups. The bits b ₁ , b ₂ , b ₃ to which the highest level of protection is to be given, including the most important information, are scrambled by the first scrambler 111; Including relevant information for the second bit to be the level of the second protection granted (b _4, b ₅₎ are scrambled by a second scrambler 112; The bits b ₆ , b ₇ , b ₈ to be given the lowest level of protection, including non-critical information, are scrambled by the third scrambler 113 (scrambling is usually performed on the serial bit flow. Although not explicitly shown in FIG. 1, the scramblers 111, 112, and 113 may be assumed to parallel-serial convert their respective input bits prior to scrambling and then serial-parallel convert again).

The scrambled 8 bits are applied in parallel to the above-described four-dimensional mapper 121, for example, identifying the four-dimensional signal point to be output using the above-described bit allocation scheme. The mapper 121 may be realized using, for example, a table look-up. At this time, the conventional passband shaping and TV modulation are performed by the TV modulator 151 of the passband shaping machine 141, respectively. The analog TV signal generated accordingly is broadcasted through the antenna 152.

Referring to the receiver of FIG. 3, an analog TV signal is received by an antenna 301, which is processed by the processing unit 311 through a conventional TV preprocessing, including demodulation, for example, an A / D converter. Is converted to digital form by 312. The signal is then equalized by passband channel equalizer 321 and applied to detector 331. The detector stores information relating to mapping, in particular information indicating the position of signal points in coordinates and information on how the signal points are divided into groups and subgroups, and so-called "slicing" operations on equalized signals. By determining how much the transmitted signal point responds to the stored information. The detector is standard, except that the coordinates of FIGS. 5 and 9 in accordance with the invention have an understanding as to how they are constructed in accordance with the invention.

The 8-bit words output by the detector 331 are descrambled by the descramblers 341, 342 and 343 respectively performing the inverse function of the scramblers 111, 112 and 113 in the transmitter. Thus, for example, a TV signal formatted to be displayable by a CRT display is generated from the descrambler output by an image signal generator 353. The TV signal is then applied to the CRT display 360.

One or more steps of advancement in an efficient C-R design can be achieved by adding redundancy to the signal coordinates. Adding redundancy allows the use of forward-error-correction coding, such as trellis coding. One of the problems with lattice coding is that the effect on the error rate of individual bits is not well known. The published work seems to focus only on the possibility of error mapping that is not related to the bit error rate of lattice coding scramble. Nevertheless, even a simple example can show how powerful C-R mapping can be obtained by using lattice coding.

Send 3 bits per signal point and assume that one of the 3 bits requires greater protection than the other bits. In a non-lattice coding system, this is done using C-R mapping with signal coordinates (1, 2; 3) as shown in FIG. The most significant bit defines the upper or lower half plane, and the other two bits define one of four possible points in each half plane. It is easy to prove that the most significant bit has a 7dB greater margin for noise than the other 2 bits. Assume that an independent one-dimensional grid code is used along each axis of FIG. 10 to actually increase the distance between rows and independently increase the distance between points in a row. Thus, the signal coordinates shown in FIG. 11 are obtained. In particular, one of the three bits is lattice coded to be two bits that select one of the four paths of FIG. 11, and the other two bits are lattice coded independently of the first bit to select one of the eight columns of FIG. 3 bits.

FIG. 2 shows a block diagram of a transmitter using the coordinates of FIG. The TV signal source 201 generates an analog TV signal that is applied to the signal source coder 204. The source coder generates a digital signal consisting of three bit binary data words (c ₁ , c ₂ , c ₃ ), assuming that bit (c ₁ ) is more important than the other two bits (c ₂ , c ₃ ). do. Bit c ₁ is scrambled by first scrambler 211, while bits c ₂ and c ₃ are scrambled by second scrambler 212.

The output of the scrambler 211 is lattice encoded by the quadrature phase grating coder 215, while the output of the scrambler 212 is lattice encoded by the in-phase grating coder 216. The 2-bit output of the grid coder 212 identifies one of the four furnaces of the coordinates of FIG. 11 as described above. These two bits are applied to a quadrature phase one-dimensional mapper 221 which generates an output that identifies one of four y-axis coordinates ± 1, ± 3. At the same time, the 3-bit output of the grid coder 216 identifies one of eight columns of coordinates in FIG. These three bits are applied to the in-phase one-dimensional mapper 222, which generates an output that identifies one of the eight x-axis ± 0.5, 1.5, ± 2.5 and ± 3.5. Thus, conventional passband shaping is performed by a quadrature-phase passband shaper 241 and an in-phase passband shaper 242, where the two outputs of the shaper are combined in an adder 243. As a result, a combined signal is applied to the TV modulator 251, where the output analog signal of the modulator is broadcast via the antenna 252.

The specific receiver for the signal generated by the transmitter of FIG. 2 is not shown. However, one of ordinary skill in the art, in this case, even if the detector stage preferably includes a maximum extreme, i.e. a Viterbi decoder, to take advantage of the coding gain provided by the lattice code, It would be easy to design the receiver using standard building blocks similar to those used in the figure.

In this kind of lattice coding, the noise sensitivity for all bits can be reduced by 3 dB. So for the error potential of 10 ^-6 , the most significant bit needs about 11 dB of SNR and the other two bits need about 18 dB of SNR (for simplicity, we assume that the channel has a flat amplitude response here). do). If a standard 8 point uncoded signal and a 16 point grid coded signal coordinate were used instead, the SNR conditions would be 18 dB and 15 dB, respectively, for the same error rate. The design for lattice coded CR mapping that is performed directly in multidimensional space will be more robust.

The foregoing description merely illustrates the principles of the invention. For example, the present invention has been described herein in connection with a digital TV transmission system. However, the present invention can be equally applied to other types of digital transmission systems. In addition, although specific coordinates are shown herein, many other coordinates of any desired dimension may be used. For example, various configurations of two-dimensional C-R mapping used to provide four-dimensional high-dimensional mapping may be used at non-uniform ratios. Alternatively, signal coordinates having different numbers of points may also be used at successive signal point intervals. All of these possibilities provide great flexibility in the design of valid multidimensional C-R mappings. In addition, since the vertical and horizontal polarization can be used at the same time, four dimensions can naturally be used in the field of HDTV. Theoretically, this allows for the simultaneous transmission of two independent QAM signals. Therefore, in such an application, there is an opportunity to realize multidimensional C-R mapping in both time (over different signal point periods) and space (between polarities).

Also, in the above embodiments, although some form of signal source coding has been used, various other methods of digital representation of TV signals, that is, other forms of signal source coding, may be used to provide high protection to some of the transmitted bits. have. Such methods include, for example, the use of trellis / convolutional codes, BCH codes, Reed-Solomon coders and combinations thereof. Unfortunately, some channel mapping schemes may extend the bandwidth of the transmitted signal or may have potential synchronization problems and may not be cost effective. In either case, if this problem can be solved, the present invention can always be combined with these schemes since any of the above schemes do not operate on the bit flow. It is also possible for the signal source coding to include other forms of processing, such as any of various forms of TV signal compression.

Also, although illustrated herein as realized by separate functional building blocks (eg, signal source coders, scramblers, etc.), the functionality of any one or more of these building blocks may be performed using one or more methods. have.

While not explicitly shown or described herein, it will be apparent to one of ordinary skill in the art that the principles of the invention may be embodied and that many other alternative devices may be devised within the spirit and scope of the invention.

Claims (19)

CLAIMS What is claimed is: 1. A method of communicating television (TV) information, comprising: generating a digital signal indicative of the television information, the digital signal consisting of at least first and second data flows of data elements; Channel mapping the digital signal; Transmitting the mapped signal on a communication channel, wherein the mapping step is such that a channel induction error for a data element of the first data flow is likely to be a channel induction error for a data element of a second data flow. Less than the possibility of the mapping step, wherein the mapping step is performed to indicate the data element so that a minimum distance between signal points representing different values of the data element of the first data flow indicates a different value of the data element of the second data flow. And selecting a sequence of signal points from the predetermined coordinates of the signal points that are greater than the minimum distance between the signal points to display. The method of claim 1, wherein said mapping step comprises lattice coding a digital signal. A method according to claim 1, wherein the coordinates are N-dimensional coordinates (N≥2). 2. The method of claim 1, wherein generating said digital signal comprises receiving said information and source coding said information using a predetermined source code. 5. The method of claim 4, wherein the digital generation step comprises processing the source coded information using at least one first predetermined processing algorithm, wherein the processing step is performed on a data element of the second data flow. And the data element of the first data flow is irrespective of the processing executed. A device for use in a digital communication system in the form in which a signal point from a predetermined signal point coordinate representing each associated data word consisting of individual data elements is communicated from a transmitter to a receiver via a communication channel, the device being each Responsive to at least one data element of the data elements of the data word, identifying whether a particular group of groups of signal points includes a signal point associated with the data word, and identifying at least one other data element of the data elements of the data word. In response, means for identifying whether a particular subgroup of the subgroups in the particular group includes a signal point associated with the data word, wherein the coordinates are divided into groups of signal points that are each divided into subgroups of signal points. The group and the subgroup are sent to which group Doeeojim emitter whether the point the receiver incorrectly determine likely transmit signal arranged to be smaller than the possibility that the receiver incorrectly determined from either the sub-group) and; Means for generating a signal indicative of a signal point from the identified subgroup and applying the generated signal to the communication channel. 7. The apparatus of claim 6 wherein the data element identifying the particular group indicates more important information than the data element identifying the particular subgroup. 7. The apparatus of claim 6 wherein the information is television signal information. 7. The apparatus of claim 6 wherein the data word is a grid encoded data word. 7. The apparatus of claim 6 wherein the coordinates are N-dimensional coordinates (N > 2). a) an apparatus for displaying television information and for generating a digital signal comprised of at least first and second data flows of data elements; b) a channel mapping channel of the digital signal using the predetermined signal coordinates, wherein the mapping is such that the circuit distance between the signal points representing different values of the data elements of the first data stream is representative of the data elements; A selection of a sequence of signal points from predetermined coordinates of the signal points intended to be greater than a minimum distance between signal points representing different values of data elements of the data flow; C) An apparatus for transmitting a signal mapped to a receiver over a communication channel, wherein the mapping is such that the likelihood of channel induced errors for data elements of the first data flow is likely to be channel induced errors for data elements of the second data flow. 10. An apparatus used in a receiver for receiving television information communicated to the receiver by a transmitter, the apparatus comprising: means for receiving a transmitted signal; Means for storing information relating to the signal coordinates and for recovering the information from the received signal in response to the stored information indicative of the location of the signal point of the coordinates. . 12. The apparatus of claim 11, wherein said mapping comprises lattice coding of a digital signal and said recovery means comprises a maximum-likelihood decoder. 13. The television information according to claim 12, wherein said mapping comprises the selection of a sequence of signal points from predetermined coordinates of a signal point to indicate a data element and wherein said stored information indicates the position of a signal point of said coordinates. Receiver device for receiving. 14. The apparatus of claim 13, wherein the coordinates are N-dimensional coordinates (N > 2). In order to indicate each associated data word consisting of individual data elements, a) a particular group of signal points associated with the data word in response to at least one data element of the data elements of each said data word Identify whether a signal point includes a signal point, and in response to at least one other data element of the data elements of the data word, identify whether a particular subgroup of the subgroups in the particular group includes a signal point associated with the data word and ; b) generates a signal indicative of a signal point from the identified subgroup and applies the generated signal to the communication channel (the group and the subgroup are unlikely to be transmitted by the receiver which may incorrectly determine from which group the transmitted signal point is from) The signal point selected from the selected signal point coordinates divided into a group of signal points each divided into subgroups of the selected signal point. A method for use in a receiver in a digital communication system in the form of being communicated from a transmitter to a receiver, the method comprising the steps of: receiving a transmitted signal point; Recovering the data word indicated by the received signal point from the received signal point (the recovery step being stored in the receiver in response to information regarding how the coordinates and the coordinates are divided into the groups and subgroups) To be executed in a digital communication system. 16. The apparatus of claim 15, wherein the data element used to identify the particular group indicates more important information than the data element identifying the particular subgroup. 16. An apparatus according to claim 15, wherein said information is television signal information. 16. The apparatus of claim 15 wherein the data word is a grid encoded data word. 16. The apparatus of claim 15, wherein said coordinates are N-dimensional coordinates (N > 2).